Positions on Mind and Body

The major positions in philosophy of mind are often presented as a ladder from substance dualism to eliminative materialism, with each rung accepting more of a physical worldview. That picture is convenient and misleading. The views answer different questions: what substances exist, whether mental properties reduce, what fixes mental kinds, whether consciousness is fundamental, and how mental causes work. A functionalist may be a physicalist or a property dualist; a panpsychist may be a physicalist in a liberal sense or reject physicalism as usually defined; a token identity theorist may deny type reduction. The labels are coordinates, not mutually exclusive boxes.

This page maps the positions against the constraint set: mental reality, mental-physical distinctness, mental causation, and physical causal completeness. Its aim is not to choose a winner but to identify where each view spends its explanatory cost.

Comparative map

The same evidence can favour different rows under different criteria. Tight brain dependence supports every view on which physical organisation fixes or lawfully controls mentality, including property dualism. Multiple realisation weighs against a simple type identity but supports both functionalism and some non-reductive views. Phenomenal conceivability arguments challenge reductive physicalism but do not by themselves choose among property dualism, panpsychism, and Russellian monism.

Substance dualism

Substance dualism holds that the subject of experience is not identical with the organism or any physical part of it. Descartes' version treats mind as essentially thinking and body as essentially extended. Contemporary versions need not accept every Cartesian doctrine; they may appeal instead to the unity of consciousness, modal arguments, or the apparent possibility of disembodied survival.

Its advantage is directness. Phenomenal and rational properties are irreducible because their bearer is not physical. Mental causation is genuine if minds interact with bodies. The same commitments generate the objections. If two qualitatively identical bodies are present, what pairs one mind with one body rather than the other? How does a non-physical cause enter a physical history without detectable incompleteness? Why do specific neural alterations modify or destroy specific capacities? An interactionist can posit psychophysical laws, but laws state the regularity whose basis the objection asks to understand.

Substance dualism must also distinguish a subject from a bare point of view. Memory, character, cognition, sensory perspective, and agency depend on bodily organisation. Removing all of them may preserve numerical identity only by leaving too little to count as the person whose survival was at issue. This connects, without settling, the afterlife and persistence debates.

Property dualism

Property dualism accepts one kind of concrete bearer - ordinarily the organism or brain - while holding that it has mental properties not reducible to physical ones. The view is attractive when the main resistance to physicalism concerns qualia rather than a separable soul. It fits brain dependence: neural states may nomologically necessitate phenomenal states without being identical with them.

The causal dilemma is severe. If the irreducible property causes behaviour, physical causal completeness appears false or effects are systematically overdetermined. If it does not, phenomenal properties are epiphenomenal: caused by physical processes but causally inert. Then reports about experience are produced physically without the experience influencing the report. The resulting match between report and experience looks accidental, and the evolutionary value of consciousness becomes obscure.

Some property dualists accept fundamental psychophysical laws; others treat phenomenal properties as strongly emergent. Both preserve the datum by adding structure at the point where the physicalist seeks derivation. The issue is whether this is an honest recognition of fundamentality or a label for the missing explanation.

Behaviourism

Methodological behaviourism restricts scientific psychology to publicly observable relations among stimulus, behaviour, and environment. Logical behaviourism makes a stronger philosophical claim: statements about mental states can be analysed as statements about behaviour or dispositions to behave. To believe that it will rain is, schematically, to be disposed to carry an umbrella, answer rain-related questions, and act in indefinitely many appropriate ways.

The position opposes the picture of a private inner object known only by introspection, and it makes evidence for mentality public. Its difficulty is that dispositions are both defeasible and mentally entangled. A person in pain may suppress every sign; an actor may display them without pain. What behaviour manifests a belief depends on desires and other beliefs, so analysing each state requires the network it was meant to reduce. Behaviour remains indispensable evidence after logical behaviourism fails, but evidence for a state need not constitute the state.

Identity theories

The type identity theory identifies mental kinds with neural kinds, as temperature is identified with mean molecular kinetic energy in an ideal gas. It offers a clean answer to causation: a mental cause and its physical cause are one event under different descriptions. Scientific identities may be discovered a posteriori, so the fact that pain and neural activity seem conceptually different does not refute identity.

The classic objection is multiple realisation. If pain can occur in humans, cephalopods, extraterrestrials, and artificial systems with physically diverse structures, no single neural type is pain. A type theorist may restrict identities to species or construct a disjunction of realisers, but the resulting physical kind seems to borrow its unity from the mental role.

Token identity requires only that each particular mental event be some physical event. It accommodates diverse realisers and physical completeness. It leaves open why events grouped by belief, pain, or intention form explanatory kinds, and whether those mental properties themselves make a causal difference. Token identity is therefore compatible with positions as different as anomalous monism and some forms of functionalism.

Functionalism

Functionalism individuates mental states by what they do: their typical causes, relations to other states, and effects on behaviour. Pain is not defined by a material but by a place in a causal organisation involving damage, attention, aversion, learning, and withdrawal. The role can be realised differently, which explains why mentality may cross biological and artificial substrates.

This is functionalism's central achievement and the basis of much cognitive science. It also creates two opposite risks. A role described too abstractly may count a nation, lookup table, or accidental system as a mind: liberalism. A role tied too closely to human organisation may exclude plausible animal or alien minds: chauvinism.

Phenomenal objections ask whether role fixes experience. An inverted spectrum holds functional relations constant while swapping colour character; an absent-qualia case holds them constant while removing experience. If such cases are possible, functional organisation is insufficient for phenomenal consciousness even if it captures cognition. A functionalist may deny their possibility, add a representational or biological constraint, or restrict the theory to some mental kinds.

Functionalism is not automatically materialism. A physical functionalist identifies realisers with physical systems; a dualist could hold that a non-physical state occupies the role. Nor does substrate independence mean that every implementation is equivalent: causal topology, temporal dynamics, embodiment, and counterfactual dispositions may be parts of the role.

Non-reductive physicalism

Non-reductive physicalism aims to preserve three claims: every concrete particular is physical, mental properties are real and not reducible to physical types, and mental explanations are causally and scientifically legitimate. Mental properties are realised by or grounded in physical organisation, much as biological or computational properties are.

The view fits ordinary explanatory practice. A program's error can explain an output although every transition is electronic; a recession can cause policy although every participant is physical. Explanations at one level need not compete with mechanisms at another. The ontology of levels supports this anti-competition model.

Kim's exclusion argument asks whether the analogy preserves distinct mental properties. If the physical realiser is sufficient for the effect, either the mental property is identical with or reducible to the causally relevant physical property, or it contributes nothing. Appeals to realisation, intervention, and proportionality try to show that higher-level causes are not rival events but apt variables in a causal model. The remaining dispute is whether that secures causal efficacy or only explanatory usefulness.

Eliminativism

Eliminative materialism treats at least some familiar mental categories as parts of a false theory rather than as permanent features of mature science. "Belief" and "desire" might go the way of phlogiston: useful in an earlier explanatory scheme but absent from the improved ontology. More modest revisionism expects categories to be transformed rather than wholly removed.

The historical possibility of elimination is real; not every successful everyday category marks a scientific kind. Yet folk-psychological explanation predicts and rationalises behaviour with unusual flexibility, and an eliminativist appears to state a belief that there are no beliefs. The self-refutation charge can be resisted by treating assertion in non-doxastic terms, but the replacement theory must then explain communication, inference, and error at least as well as the rejected vocabulary.

Consciousness creates a distinct limit. A theory may eliminate the idea that experience has intrinsic, ineffable qualia while explaining why subjects judge that it does. It is harder simply to deny that anything appears any way, because the seeming to seem is itself the target phenomenon. Illusionism is therefore better understood as a specific revisionary theory of phenomenal properties than as the claim that nobody is conscious in any sense.

Panpsychism and monist alternatives

Panpsychism holds that experiential or proto-experiential features are fundamental and widespread. It changes the explanatory task: consciousness need not emerge from a wholly non-experiential base, so no absolute jump occurs. The price is the combination problem. Even if microphysical entities have rudimentary experience, why and how do many such subjects compose one unified subject rather than merely a collection?

Russellian monism begins from a different observation: physics characterises entities structurally and relationally - through dispositions, equations, and causal roles - while saying little about their intrinsic categorical nature. That hidden nature might ground consciousness. Neutral monism likewise proposes a base that is in itself neither mental nor physical, with both domains arising as organisations or aspects of it.

These views avoid interaction between independent substances because mind and matter share a basis. Their weakness is positive specification. If the neutral or categorical base is described only as whatever makes both sides possible, it inherits the explanatory gap under a new name. If it is described as already experiential, the combination problem returns.

Cross-cutting choices

A complete theory needs answers on several axes, not just one row of the table:

This explains why apparent hybrids are often coherent. One may combine token identity, functionalism about belief, representationalism about perception, and Russellian monism about phenomenal consciousness. Eclecticism is not automatically a vice: different mental phenomena may form different natural kinds. But each additional account creates interface questions, such as how functionally individuated beliefs interact with irreducible experiences.

What would discriminate among the views?

No single observation announces that functionalism or dualism is true. Evidence can nevertheless discriminate when paired with independent predictions.

• If a proposed mental kind generalises across radically different realisers while preserving causal organisation, simple type identity loses support.
• If phenomenal reports can be exhaustively explained by mechanisms that make no reference to the properties reports seem to describe, illusionism gains a resource; whether the phenomenal target has been explained remains disputed.
• If a theory of consciousness assigns sharply different states to systems that can be experimentally compared, intervention and dissociation can favour one mechanism.
• If purported non-physical causation produces no difference from a physically complete account, interactionism loses explanatory necessity, although absence of a measured discrepancy is not a logical proof of closure.

The crucial phrase is "paired with independent predictions." Re-labelling every possible result after the fact protects a view from evidence by depriving it of content. The next page develops the methodological standards needed to avoid both crude scientism and insulation from science.

Selected references

• Block, Ned. "Troubles with Functionalism" (1978).
• Churchland, Paul M. Matter and Consciousness (1984).
• Kim, Jaegwon. Physicalism, or Something Near Enough (2005).
• Place, U. T. "Is Consciousness a Brain Process?" (1956).
• Putnam, Hilary. "The Nature of Mental States" (1967).
• Russell, Bertrand. The Analysis of Matter (1927).